Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
HP8640B -140/+10dBm instead of -130/+20dBm?
|
I'm new to this group. I've diligently searched the archives but can find no mention of early 8640B models seemingly intended to produce +10dBm maximum output instead of the more usual +20dBm.
I have a pair of these 8640B signal generators, the early-model one having been recently purchased so I can make IMD measurements on my homebrew ham radio receivers. The other one I've had for some years has the common fault with broken plastic gears and Paul K is helping me out with this, hence in the meantime me putting the later purchase into service and being "surprised" at the lower output. The serial number is 1434U00265, so it's a 1974 model. The output range switch has a scale marking at -140dBm and also one at +20dBm, but the switch does not rotate beyond +10dBm. Does anyone know when HP increased the power output? Does it mean this example doesn't contain the dreaded monolithic output amplifier? Although the range/deviation switches use proper wafers instead of microscopic spring fingers on plastic discs the mechanically mad plastic gears are there so I suspect there will be service work to do in the future. Is it OK to ask applications questions on here? I'm wondering how much reverse isolation I need between my pair of generators to make accurate IMD measurements? I'll be looking for spurs at the noise floor of the receiver, say -130dBm or lower, and for a decent receiver this will need -30dBm signals from each generator. I'm thinking about, say, a 10dB pad between my hybrid combiner and receiver, with maybe a 20dB pad between each generator and the hybrid. Is this enough to ensure there will be no IMD from the generators caused by the crosstalk between them? Or do I need to think about external amplifiers and yet more attenuators? Thanks, Alan G3XAQ |
|
Hello,
On Thu, Oct 22, 2015 at 12:30 PM, Alan Ibbetson alan@... [hp_agilent_equipment] <hp_agilent_equipment@...> wrote: I have both models, the output level range is basicly the same on both, but the one with the amplitude dial with the -140 dBm mark can go to +20 dBm too using the fine level knob. You'll notice that the level full scale will change as you move the fine level knob and a small mark under the large dBm marks moves to indicate the actual output power. HTH Frank IZ8DWF |
|
For IMD measurements, you can use any coupler that can couple 2 or more generators together with some isolation.? 6dB Hybrid combiners are quite popular and broadband. Same goes for resistive couplers.? A lot of interesting information on the different couplers can be found here:? My personal favorite is a Wilkinson splitter. I do most of the measurements on VHF and UHF and with so high levels that a ferrite based coupler wil generate excessive IMD. The isolation between the generators is normaly not a problem, with the 8640's you will most of the time use 20 or 30dB attenutation on both generators, leading to better than 60dB isolation, even without any isolation in the coupler.? When measuring on receivers the input signals will often be low, but if measuring IMD on a power amplifier chain, then you may end up needing some external amplifiers that goes after the generators. A set of good step attenuators after the combiner helps in setting the step level, and determining if the IMD is generated in the coupler or after it. The book "Experimental methods in RF design" by Hayward, Campbell and Larkin have a good treatment of most IMD measurement setups. BR. Thomas LA3PNA. ? 2015-10-22 14:49 GMT+02:00 Alan Ibbetson alan@... [hp_agilent_equipment] <hp_agilent_equipment@...>:
--
?Please? avoid sending? me? Word? or? PowerPoint? attachments. ?See? <> PDF is an better alternative and there are always LaTeX! |
|
Hello Thomas. Thanks very much for those useful references. It seems I won't be needing external amplifiers for my simple receiver testing. Jack at Clifton Labs mentions using LPFs to avoid confusion caused by generator harmonics, but mine are -60dBc so I hope added filtering will not be necessary.
Thanks everyone. Best wishes, Alan G3XAQ -- Alan Ibbetson alan@... |
|
There is an extensive HP application note on using the 8640 specific to IMD. Significant improvements in IMD measurement is realized but not allowing a pair of 8640 generators to modulate each other via their internal ALC loops. Hence, HP provided the ability to TURN THE GENERATOR ALC LOOP OFF. If you pop the hood on the 8640 your should find an ALC switch. Of course, I am going out on a limb and cannot recall if early units had this option, however I do believe the date code your addressing will work this option. On Thursday, October 22, 2015 4:35 PM, "Alan Ibbetson alan@... [hp_agilent_equipment]" wrote:
?
Hello Thomas. Thanks very much for those useful references. It seems I
won't be needing external amplifiers for my simple receiver testing. Jack at Clifton Labs mentions using LPFs to avoid confusion caused by generator harmonics, but mine are -60dBc so I hope added filtering will not be necessary. Thanks everyone. Best wishes, Alan G3XAQ -- Alan Ibbetson alan@... |
|
Yes there is.
toggle quoted message
Show quoted text
, I believe. On Oct 23, 2015, at 10:16 AM, alan victor amvictor88@... [hp_agilent_equipment] wrote:
|
|
On Figure 16, page 10, the traces are mislabeled. The black traces are the 608, while the blue traces are the 8640.
toggle quoted message
Show quoted text
JFYI, Dave Wise -----Original Message-----
From: hp_agilent_equipment@... [mailto:hp_agilent_equipment@...] Sent: Thursday, October 22, 2015 2:31 PM To: hp_agilent_equipment@... Subject: Re: [hp_agilent_equipment] HP8640B -140/+10dBm instead of -130/+20dBm? Yes there is. , I believe. On Oct 23, 2015, at 10:16 AM, alan victor amvictor88@... [hp_agilent_equipment] wrote:
------------------------------------ ------------------------------------ ------------------------------------ Yahoo Groups Links |
|
Thanks, guys. I see the ALC on/off switch S1 on board A26A4 in the manual but it involves opening the diecast box and maybe pulling the board out to toggle the switch, though it's at the top edge of the board and perhaps can be moved with a pointed stick.
The app note fig 14 suggests a hybrid with 40dB isolation will achieve the same result as disabling the ALC. My Hatfield Instruments 6dB hybrid shows 50dB isolation at 14MHz so I am inclined to go that route. Assuming everything is well behaved one might hope that keeping the generators at -10dBm (or maybe as high as 0dBm) would give spurs at least -100dBc, which is the minimum I need for evaluating 100dB DR receivers, and still give me 14dB or more attenuation to dial in with a step attenuator with receiver noise floors around -130dBm. Does anyone know just how low the IPs go if both the ALC is disabled and a combiner are used? Do my numbers look realistic, or have I made some idiot arithmetical error? Thanks, Alan |
|
Dave Brown
Alan-
toggle quoted message
Show quoted text
Not used 8640s for many years so can't answer your query, sorry. Your figures make sense to me, BTW. But check the hybrid isolation at the highest frequency you're going to test at. If you're looking at SDR rx testing, as opposed to conventional rx architectures- you might like to check this out- implementation depends on whether you can get hold of some old white noise gear. 73 Dave, ZL3FJ ----- Original Message -----
From: "alan@... [hp_agilent_equipment]" <hp_agilent_equipment@...> To: <hp_agilent_equipment@...> Sent: Friday, October 23, 2015 11:21 AM Subject: Re: [hp_agilent_equipment] HP8640B -140/+10dBm instead of -130/+20dBm? Thanks, guys. I see the ALC on/off switch S1 on board A26A4 in the manual but it involves opening the diecast box and maybe pulling the board out to toggle the switch, though it's at the top edge of the board and perhaps can be moved with a pointed stick. |
|
Andy Gardner
Thanks Dave, I've annotated my paper copy.
toggle quoted message
Show quoted text
Cheers, On 23/10/2015, at 11:08 AM, David Wise david_wise@... [hp_agilent_equipment] wrote:
On Figure 16, page 10, the traces are mislabeled. The black traces are the 608, while the blue traces are the 8640. |
|
A friend of mine used to work as a design engineer for HP at Queensferry in the UK, though he didn't himself work on the 8640B. I asked him about turning off the ALC versus using external attenuators and couplers. He said
--- Turning off the ALC fixed ALC pumping. It doesn¡¯t fix the diode nonlinearity from the detector influencing the output by being a non-linear load, nor the non-linearity of the output amp. Yes, a coupler with >=40dB isolation would be good. Snag is that broadband HF ones use transformers, and to get higher frequency coverage drives the designers to scoddy little cores. Ferrites do intermod too, so finding a low intermod coupler is difficult. --- The Clifton Labs website that Thomas referenced has details of couplers that don't have significant IMD so that is the path I am going to choose, leaving the inconvenience of ALC disabling for the day when I can build receivers with substantially more than 100dB IMD dynamic range. Thanks, Alan G3XAQ |
|
The ALC pulling or modulation of the loop created IM in itself, hence it is worth doing. I know because I have measured the condition for both cases. There is little one can do because of the PIN attenuator system but that is a secondary IM contribution. On the 40's I have it is straight forward to accomplish, may be different in earlier models. On Friday, October 23, 2015 9:20 AM, "Alan Ibbetson alan@... [hp_agilent_equipment]" wrote:
?
A friend of mine used to work as a design engineer for HP at Queensferry
in the UK, though he didn't himself work on the 8640B. I asked him about turning off the ALC versus using external attenuators and couplers. He said --- Turning off the ALC fixed ALC pumping. It doesn¡¯t fix the diode nonlinearity from the detector influencing the output by being a non-linear load, nor the non-linearity of the output amp. Yes, a coupler with >=40dB isolation would be good. Snag is that broadband HF ones use transformers, and to get higher frequency coverage drives the designers to scoddy little cores. Ferrites do intermod too, so finding a low intermod coupler is difficult. --- The Clifton Labs website that Thomas referenced has details of couplers that don't have significant IMD so that is the path I am going to choose, leaving the inconvenience of ALC disabling for the day when I can build receivers with substantially more than 100dB IMD dynamic range. Thanks, Alan G3XAQ |
|
Hurray! Someone benefited from my OCD moment. :)
toggle quoted message
Show quoted text
-----Original Message-----
From: hp_agilent_equipment@... [mailto:hp_agilent_equipment@...] Sent: Friday, October 23, 2015 1:04 AM To: hp_agilent_equipment@... Subject: Re: [hp_agilent_equipment] HP8640B -140/+10dBm instead of -130/+20dBm? Thanks Dave, I've annotated my paper copy. Cheers, On 23/10/2015, at 11:08 AM, David Wise david_wise@... [hp_agilent_equipment] wrote: On Figure 16, page 10, the traces are mislabeled. The black traces are the 608, while the blue traces are the 8640. ------------------------------------ Posted by: Andy Gardner <ceo@...> ------------------------------------ ------------------------------------ Yahoo Groups Links |
to navigate to use esc to dismiss